Really, they did it to themselves. They should have flood-proofed their tents.

/s

I kind of agree, except for one big difference: cloud save sync. Currently, Heroic et al can sync game saves for Windows GOG games, but for Linux-native games (or Linux builds of multiplatform games), that’s not possible. GOG would have to support Linux cloud saves on the server side.

(For me this is Silksong.)

When they officially have Galaxy for Linux, they will finally have a reason to implement it. And at that point, presumably, the FOSS launchers will be able to utilize it too.

Other than that, having the option of an official GOG Galaxy launcher, with smoother GOG integration and active work both on the client and the server to make everything work better, is a good thing no matter how you look at it. Yes it’s late, but it’s definitely not too late.

I might keep using Heroic. I might switch to this. Time will tell.

I don’t know what to say other than your description doesn’t match the opening paragraphs on Wikipedia (which I linked earlier), which AFAIK match the common meaning of the terms. The existence of cross-compilation really muddies the water here, so to clarify, in this context when I’m talking about compilation I’m talking about compiling to machine code that can run natively on the hardware in question (CPU or GPU). Compiling to anything else (bytecode, different programming language, whatever) is irrelevant for the discussion IMO.

And interpreters do not compile to machine code. If they do, it’s called JIT compilation, not interpretation.

For short, the term Interpreter I think is misunderstood here.

Definitely, but I strongly think it’s you who is wrong here.

No, it’s not compiled on the fly. With an interpreter, there is never a sequence of CPU instructions corresponding to the source code. All of the CPU instructions that run on the CPU are from the interpreter’s executable. The source code (or sometimes bytecode, which is still not CPU instructions) is treated as data. New machine code is never created.

Compiling means creating new machine code that can run on the CPU. This happens in JIT. It doesn’t happen in an interpreter.

I’m not sure what you mean by that. Those are script languages that are shared by its source code.

Yes, that’s why I chose them.

They still need to be compiled.

No - they don’t. Sometimes they can be compiled, but when using a “pure” interpreter, there is no compilation step. The source code is read and executed on the fly. https://en.wikipedia.org/wiki/Interpreter_(computing)

“In computing, an interpreter is software that executes source code without first compiling it to machine code.”

JIT is actually a kind of middle-ground, where it does compile the source code (or sometimes just the most critical parts of it) for better performance.

Python is traditionally only interpreted AFAIK, but it’s nuanced and it’s a rabbit hole I can’t hop into right now.

So, my basic understanding is that shaders currently use a JIT compilation model: the first time a shader is needed, it is compiled first and then executed. Which obviously causes stutters. If there was a GPU shader-interpreter that could skip the compilation step at the cost of execution speed (but still be faster than fully compiling the shader) then, well, that would be what I mean.

Okay, so the way I imagine the ubershader is as an interpreter for uncompiled shaders. It may be much slower than the compiled shader, but it doesn’t need to wait. Just like JavaScript or Python or Lua can be interpreted instead of compiled. So if a shader is required for this frame but isn’t compiled yet, you use the ubershader to render it right now while asynchronously sending it to be compiled for full-speed rendering on future frames.

But yeah, it might be completely impossible, I really know nothing about how shaders work (both before and after compilation).

I started writing something in response but I quickly realized I’m also not knowledgeable enough to even discuss this further…

Interesting. My first thought upon reading this is: I wonder if anything akin to Dolphin’s Ubershaders is viable for PC games? Specifically hybrid mode. If a game implements this, it would completely solve this issue. When a shader is first needed, it will start asynchronously compiling, but the ubershader (which would be compiled at game launch) would immediately be able to render it - albeit less efficiently than the fully compiled shader. This would replace stutters with temporary framerate dips, which might not even be noticeable (since only a handful of newly-encountered shaders actually need this slower treatment in any given frame).

Heck, if this is viable, it can be implemented in the GPU (or its driver) and solve the problem for every game. That would be sweet.

I loved this game. Really awesome singleplayer campaign. This is pretty exciting!

The Arab parties’ 18 seats is the biggest game changer. The Netanyahu block will obviously never partner with them, and neither they nor the Eisenkot block is large enough to form a government coalition without them. So the only way we make it out of this mess is with some real Arab representation in the government.

I really hope that happens. But it’s worth noting that some (not all!) of the Arab parties don’t acknowledge Israel’s legitimacy, so obviously cannot be included in Israel’s government.

Please oh please let this work out. I want to get off this Bibi horror ride.

The alternative is to link to a specific instance. In which case if you open it you’ll be logged out.

Thanks.

I couldn’t care less about the downvotes, just like you I’m just interested in having a good conversation and sharing accurate information. So thank you for providing it!

You’re right about all of that. Especially the takeaway that it’s nuanced. Yeah, it really is.

I think what most people, including you, are reacting to is who is doing this rather than what is being done. And rightly so. I don’t want to trust Epic (or Mic🤮soft) with this kind of thing. But anti-cheat is nuanced, and I think I can persuade you (if you don’t agree already) that some kind of anti-cheat that actually works to some extent is needed.

You brought up:

… you have to ask yourself what platform you’re participating in and why.

I think the answer is pretty obvious here. The platform is an online game. Let’s say Fortnite just because it’s the obvious example to use (personally I’ve never played it). The reason you want to participate is for fun, for competition (maybe even professional), or both.

One thing is very clear: if other people playing Fortnite are cheating when playing with you, it prevents you from having fun or winning the competition. Unless you are a cheater yourself, you only want to play with non-cheaters.

So obviously you want there to be some way you can play Fortnite with some level of guarantee that other players aren’t cheating.

So let’s say you, the player, get to decide how this is enforced. The stipulation is that every player including you will be subject to the same level of enforcement, and any player that isn’t will be locked out of the platform. How do you do it?

You and every player will have to somehow credibly convince the game that you’re not cheating. The only way to do that is to reveal some information about the system running the game. The TPM and all those other technologies that you seem to understand much better than I do, are the way to do that credibly. The only question left, then, is which information (or “how much”) would actually be sent to the server to provide this proof.

That’s the trade-off. I agree with you that these companies will almost inevitably exploit this system to extract data about you that has nothing to do with cheating. If not at first, then soon afterwards. And that’s an endless battle that we’ll continue to fight. But it’s also besides the fundamental point.

Cheating in online games is a real problem that actually needs solutions. I think the framework presented by EAC v2 is a good one, and can be used to enforce anti-cheat on Linux without spying on you in any other way. That, I think, is a good thing.

What would be best, of course, is if a more trustworthy organization made a similar anti-cheat solution following EAC v2’s footsteps.

I didn’t because linking to a post or comment on Lemmy (or the fediverse at large) is awful. Does this work? https://vger.to/discuss.tchncs.de/comment/29001869

So are you saying it doesn’t get root type privileges? And can be as trusted to not be as invasive as anticheat is on windows being able to look through your entire system?

From my understanding, that’s right. My understanding is very shallow though, so you should look into it yourself if you really care enough. I think the links I posted are enough of a starting point.

Personally, I don’t even play any competitive multiplayer games so I generally have no interaction with anticheat one way or another. For that reason, I don’t see myself doing the steps required to get EAC v2 working.

But, given that there are loads of people who:

1. Want to play and enjoy these competitive multiplayer games
2. Don’t want cheaters to ruin the game for them
3. Acknowledge that cheaters will ruin the game if nothing is done to prevent that

Then, some trade-offs are inevitable. Anti-cheat is required for that, there’s no getting around it. That is the “good reason” for which the attestation is needed.

Maybe for you that’s not a good reason - that’s fine, it’s the same for me, I just don’t play those kinds of games and so the games I play don’t need anticheat. There’s really no reason to get angry with a solution that DOESN’T AFFECT YOU if you’re not interested in it, but actually serves a purpose for others.

This isn’t putting anything sleazy in your kernel. Please take a look at my reply to another comment.

This isn’t putting anything sleazy in your kernel. Please take a look at my reply to another comment.

That’s not what this is doing. From my understanding, the kernel doesn’t include any new code made by Epic. It just needs to support Secure Boot and a bunch of other requirements, all of which are already possible with existing Linux solutions.

The anti-cheat wants to verify that you haven’t fucked with your system, there are ways to do that and EAC requires that your setup can credibly after attest to it.

Video going over the requirements (with timestamp): https://youtu.be/3Xi7m0CYk5s?t=7m39s

Source: https://onlineservices.epicgames.com/news/introducing-easy-anti-cheat-linux-v2

On Windows, we rely heavily on a kernel driver to protect the games. This driver isn’t always on, it starts when a game is launched and stops when the player exits. For EAC Linux v2 we will not leverage an equivalent kernel module, and instead we’ll take full advantage of existing Linux platform security features like measured boot, the kernel lockdown LSM, and IOMMU support, plus our own eBPF logic to maintain integrity and detect cheaters. These security features are not available on every distro and may require configuration changes to activate.

I agree with your sentiment though. But somehow, against all odds, the “programming is too hard” people landed on a solution that sounds pretty good to me.

I agree about leap minute, I really don’t see why they leaped (ha) straight to an hour.

Well, I kinda do - working with whole-hour offsets is kind of a solved issue (timezones, DST) whereas there isn’t really a solution for a minute. The clever people at Google (IIRC) came up with the “smear second” as a pretty elegant way to implement a leap second, but the same approach wouldn’t scale all that well to a full minute. You’d have to smear it over 2 months or something.

But if the time does come to implement a leap hour, that’s going to be just as hard to deal with! You basically have to redefine UTC to be what was previously UTC+1:00, and then what? Computers can’t deal with that any more than that can deal with 61 seconds in a minute.

“We already have daylight saving where we change by an hour and no one gets bothered about that,”

What the fuck. EVERYONE gets bothered about that!

I can actually get behind this leap hour idea, but holy crap that is a bad way to present it